The MICCAI Federated Tumor Segmentation (FeTS) Challenge 2024: Efficient and Robust Aggregation Methods

We present the design and results of the MICCAI Federated Tumor Segmentation (FeTS) Challenge 2024, which focuses on federated learning (FL) for glioma sub-region segmentation in multi-parametric MRI scans. Unlike previous FeTS challenges, this iteration exclusively evaluates novel weight aggregation methods for increased robustness and efficiency. Participating methods from six teams are evaluated using a standardized FL setup and a multi-institutional dataset derived from the BraTS glioma benchmark—a dataset consisting of 1,251 training cases, 219 validation cases, and 570 hidden test cases, with segmentations of enhancing tumor (ET), tumor core (TC), and whole tumor (WT). Teams are ranked by a cumulative scoring system that accounts for segmentation performance—measured by Dice Similarity Coefficient (DSC) and 95th percentile Hausdorff Distance (HD95)—and communication efficiency assessed through the convergence score. A PID-controller-based approach emerges as the top-performing method, achieving a mean DSC of 0.733, 0.761, and 0.751 for ET, TC, and WT, respectively, with corresponding HD95 values of 33.922mm, 33.623mm, and 32.309mm, while also being the most efficient with a convergence score of 0.764. These results contribute to ongoing advances in FL, building on top-performers from previous iterations of the challenge and surpassing them, highlighting the potential of PID controllers as a powerful mechanism for stabilizing and optimizing weight aggregation in FL. The challenge code is available at <a href='https://github.com/FeTS-AI/Challenge'>https://github.com/FeTS-AI/Challenge</a>

Authors

Institutions

Publication Details

Journal
The Journal of Machine Learning for Biomedical Imaging
Published
2025-12-05
DOI
https://doi.org/10.59275/j.melba.2025-5242
Citations
2
Primary Topic
Glioma Diagnosis and Treatment
Type
article
Field-Weighted Citation Impact
3.25
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

The MICCAI Federated Tumor Segmentation (FeTS) Challenge 2024: Efficient and Robust Aggregation Methods

Bjoern Menze, Kevin Ta, Jonathan Passerat‐Palmbach, Suprosanna Shit et al.
2 citations
The Journal of Machine Learning for Biomedical Imaging
Glioma Diagnosis and Treatment
3.25
article

The MICCAI Federated Tumor Segmentation (FeTS) Challenge 2024: Efficient and Robust Aggregation Methods

Bjoern Menze, Kevin Ta, Jonathan Passerat‐Palmbach, Suprosanna Shit, Akis Linardos, Ujjwal Baid, Suleiman A. Khan, Leon Mächler, Ivan Ezhov, Dimitris P. Makris, Verena Chung, Prashant Shah, Patrick Foley, Spyridon Bakas, Mojtaba Jafaritadi, Micah Sheller, Brandon Edwards, Sarthak Pati, Daewoon Kim, Elina Kontio, Giacomo Tarroni, Johannes C. Paetzold, Manuel Nickel, Vasilis Siomos, Muhammad Irfan Khan, David Naccache, Gustav Grimberg, Leonard Klausmann
article en
2 citations

Abstract

We present the design and results of the MICCAI Federated Tumor Segmentation (FeTS) Challenge 2024, which focuses on federated learning (FL) for glioma sub-region segmentation in multi-parametric MRI scans. Unlike previous FeTS challenges, this iteration exclusively evaluates novel weight aggregation methods for increased robustness and efficiency. Participating methods from six teams are evaluated using a standardized FL setup and a multi-institutional dataset derived from the BraTS glioma benchmark—a dataset consisting of 1,251 training cases, 219 validation cases, and 570 hidden test cases, with segmentations of enhancing tumor (ET), tumor core (TC), and whole tumor (WT). Teams are ranked by a cumulative scoring system that accounts for segmentation performance—measured by Dice Similarity Coefficient (DSC) and 95th percentile Hausdorff Distance (HD95)—and communication efficiency assessed through the convergence score. A PID-controller-based approach emerges as the top-performing method, achieving a mean DSC of 0.733, 0.761, and 0.751 for ET, TC, and WT, respectively, with corresponding HD95 values of 33.922mm, 33.623mm, and 32.309mm, while also being the most efficient with a convergence score of 0.764. These results contribute to ongoing advances in FL, building on top-performers from previous iterations of the challenge and surpassing them, highlighting the potential of PID controllers as a powerful mechanism for stabilizing and optimizing weight aggregation in FL. The challenge code is available at <a href='https://github.com/FeTS-AI/Challenge'>https://github.com/FeTS-AI/Challenge</a>

The Journal of Machine Learning for Biomedical ImagingVol. 3(December 2025)
OTH Regensburg (DE), Indiana University Health (US), Sage Bionetworks (US), Intel (United States) (US), Seoul National University (KR), Turku University of Applied Sciences (FI), St George's, University of London (GB), University of Zurich (CH), Kingston University London (GB), Cornell University (US), University of Tabriz (IR), University of California San Francisco Medical Center (US), Laboratoire de Géologie de l’École Normale Supérieure (FR), Weill Cornell Medicine (US), Indiana University School of Medicine, City St George's, University of London (GB), Imperial College London (GB), Indiana University – Purdue University Indianapolis (US), Technical University of Munich (DE), Stanford University (US)
Openalex Percentile: Top 7%
Glioma Diagnosis and Treatment
3.25
DOIOpen Access PDFOpenAlex
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.